Although each method has its own advantages and disadvantages compared with the conservative mercury-in-glass thermometers, there are conflicting opinions about the optimal anatomic site for measuring body temperature as well about the variations in measurements with different methods. In this study, we aimed to assess the accuracy and reliability of measurements obtained from the axilla with the chemical thermometer (Tempa DOT TM) compared with the classic mercury-in-glass instruments. Sixty randomly selected pediatric patients who were admitted to our hospital were enrolled. Simultaneous temperature axillary measurements (n: 1300) were performed with the chemical thermometer and mercury-in-glass instruments. The mean results of the axillary mercury-in-glass thermometers and axillary chemical thermometer were 36.8 +/- 0.6 and 37.2 +/- 0.7, respectively. The Bland-Altman plot of differences suggests that 95% of the chemical thermometer (Tempa.DOT TM) readings were within limits of agreement (+0.37 and -1.24 degrees C) when mercury-in-glass thermometer is considered as the standard. Our results showed that limits of agreement were wide (+0.37 and -1.24 degrees C) between readings of axillary mercury-in-glass thermometers and chemical thermometers. Since approximately 20% of febrile patients with mercury-in-glass temperature were misdiagnosed as afebrile with measurements via chemical thermometer, we suggest that the axilla is not a suitable anatomic site for screening of fever with Tempa.DOT. Further studies involving larger study groups with similar age should be done to more definitely assess its screening value in pediatrics.